Professional Documents
Culture Documents
A R T I C L E I N F O A B S T R A C T
Keywords: Research on low temperature resistance and fracture behavior of additively manufactured (AMed) titanium alloy
Ti6Al4V alloy is lacking. In this study, both tensile and fracture behavior of Ti6Al4V alloy produced by laser powder bed fusion
Temperature effect (LPBF) were investigated at four temperatures (293, 223, 173, 123 K). Microscopic observation and finite
Fracture property
element analysis were used to study the fracture mechanism. A correlation of fracture toughness and temperature
Tensile strength
Laser powder bed fusion
between 123 and 293 K was obtained. As the temperature decreases, the fracture mode of the specimen changes
from ductile fracture to a mixed mode of ductile and brittle fracture, the tensile strength increases and the
fracture toughness decreases. These differences could be interpreted by the crack tip plastic zone (CTPZ) and
microscopic plastic deformation. Compared to the wrought Ti6Al4V alloy, the fracture originations are mainly
from LPBF defects, i.e., lack of fusion (LOF) defects and gas pores.
* Corresponding author at: State Key Laboratory of Nonlinear Mechanics (LNM), Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, China.
E-mail address: qianguian@imech.ac.cn (G. Qian).
https://doi.org/10.1016/j.tafmec.2023.103931
Received 11 March 2023; Received in revised form 7 May 2023; Accepted 9 May 2023
Available online 18 May 2023
0167-8442/© 2023 Elsevier Ltd. All rights reserved.
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
within columnar prior β grains result in a basket weave structure [17]. original cylindrical and rectangular blocks were printed from pre-
These structures have a great influence on the tensile property, mode I alloyed gas-atomized Ti6Al4V powder using the same processing pa
plane strain fracture toughness (KIC), fatigue crack growth rate, and rameters on a commercial EOS M290 laser powder bed fusion system in
unnotched fatigue strength. Xi’an. The Ti6Al4V powder particles used here were ranged from 15 to
In general, the yield strength and ultimate tensile strength of the 53 μm in diameter and have a normal powder size distribution (PSD)
LPBF Ti6Al4V alloy can meet or even exceed those of conventional with 50% of the particles smaller than 45 μm. Due to the high reactivity
fabrication, while the ductility of LPBF Ti6Al4V is generally lower than of the titanium alloy, the printing process was performed in an inert
that of forged and cast Ti6Al4V alloys [18,19]. Fatigue crack growth rate Argon atmosphere with a purity of 99.99%. The chemical composition of
ΔKth for heat treated LPBF Ti6Al4V reported in the literature is similar to the material has been tested and is listed in Table 1. The LPBF param
√̅̅̅̅̅
that of the wrought material (3.1 ~ 4.6 MPa m), except that the high eters used in this study were: laser power (P) of 330 W, hatch spacing (h)
cycle fatigue strength (ranging from 200 ~ 350 MPa) is significantly of 0.09 mm, scanning speed (v) of 2000 mm/s, and layer thickness (t) of
inferior to that of the wrought alloy (450 ~ 650 MPa) [20]. Excessive 0.03 mm, as shown in Table 2. The final input laser energy density (E) is
presence of defects, such as pores and lack of fusion (LOF), has been 60.3 J/mm3, which is calculated by the following equation [15].
associated with poor fatigue performance in certain studies. This can P
result in either internal or surface cracking [21]. The value of KIC of the E= (1)
v⋅h⋅t
as-built LPBF Ti6Al4V alloys is typically only between 16 and 28
√̅̅̅̅̅
MPa m, owing to the presence of inherent defects and residual stress The bidirectional laser beam scanning strategy adopted is "cross-
[22]. This value is much lower than that of the wrought produced hatching", as shown in Fig. 1(a). According to [30], the "cross-hatching"
√̅̅̅̅̅
Ti6Al4V alloys, which ranges from 30 to 80 MPa m depending on the scanning strategy makes the entire laser energy input more balanced
specific microstructure [20,23]. After improving the process parameters throughout the whole layers, which effectively prevents defect accu
and applying heat treatment, the KIC of LPBF Ti6Al4V alloys can increase mulation and propagation. The substrate is preheated to 35 ◦ C to reduce
√̅̅̅̅̅ the effect on the microstructure of the heat affected zone. Later, the
to 48 ~ 58 MPa m [16]. Furthermore, duplex heat treatment can
object is created by selectively solidifying of the deposited metal powder
significantly improve the KIC, with values ranging from 77 ~ 106
√̅̅̅̅̅ layers with a laser beam. After the first layer is printed, the substrate is
MPa m due to coarser α laths and preservation of the prior β meso-
dropped one layer and the laser beam is rotated 90◦ to melt the powder
structure [24]. LPBF Ti6Al4V alloy printings with large-scale configu
newly spread on the surface of the object. The above process is repeated
ration have been used for heavy and often permanent structural com
until the object is fully formed.
ponents, with the challenge of tending to fracture [25].
The round rods with a diameter of 12 mm and a length of 68 mm and
The mechanical behavior of titanium and its alloys at cryogenic
the sample blocks of 72 × 32 × 33 mm3 were printed along the vertical
temperatures has been extensively studied. However, the applicability
direction on the build platform, also known as 90◦ printing, as shown in
of Ti6Al4V alloy has been limited to 77 K due to its low cryogenic
Fig. 1(b). The coordinate system TD, RD, and AD in Fig. 1 refer to the
fracture toughness, which is insufficient for temperatures below 77 K
tangential direction, radial direction, and axial direction, respectively.
[26]. Yuri et al. [27] discovered that the fatigue strength of notched
The axial direction corresponds to the build direction (BD). After LPBF
specimens is lower at 4 K than at 77 K, primarily because of the
but prior to tensile and fracture specimen machining, the printed orig
increased notch sensitivity at lower temperatures. Singh et al. [28]
inal cylindrical and rectangular blocks were first annealed at 800 ◦ C for
investigated the plastic deformation and tensile strength of Ti6Al4V
4 hours and then cooled in an Argon atmosphere to relieve the residual
alloyed with minor additions of Boron at cryogenic temperatures, and
stresses produced in the LPBF process. The Archimedes method was used
the yield strength variation at different temperatures shows that the
to determine the relative densities of the test specimens and the test
colony size determines the yield strength of the alloy at 77 K, implying
result is 99.33%.
that dislocation-mediated plasticity dominates above 77 K, while twin
ning dominates the flow response at 20 K. Up to now, it is clear that there
is a lack of studies on the tensile and fracture behavior of LPBF Ti6Al4V 2.2. Specimens and mechanical tests
parts at low temperatures. Considering the wide application of LPBF
Ti6Al4V titanium alloy in aerospace and space exploration, it is of great Electron backscattering diffraction (EBSD) and optical microscopy
interest to investigate the temperature dependence of the tensile and (OM) were used to characterize the initial microstructure of the LPBF
fracture properties. Ti6Al4V alloy. The samples observed were mechanically ground with
This paper aims to investigate the effect of temperature on the uni SiC sandpaper (320–5000 particle size) and polished with 0.05 μm Silica
axial tensile and fracture behavior of LPBF Ti6Al4V alloys. A series of suspension (SiO2 and H2O2). The grain structure was characterized by
quasi-static tensile tests and fracture toughness experiments were con EBSD technique in a scanning electron microscope (ZEISS SIGMA 300).
ducted at different temperatures. The initial microstructure was char The scanning step was 0.5 μm and the scanning area was 300 × 400 μm2.
acterized using the electron backscatter diffraction (EBSD) analysis. The Then, HKL Channel 5 software was used for microstructure data anal
fracture surfaces were observed by scanning electron microscopy (SEM) ysis. The polished samples were etched with Kroll reagent (100 mL
and the fracture mechanisms of the material were discussed. In addition, distilled water, 5 mL Nitric acid, and 2 mL HF) for about 10 s, and
finite element (FE) simulation was used to calculate the size of the crack examined with an Olympus BX51 optical microscope (OM) for metal
tip plastic zone (CTPZ), which is the area of local plastic deformation of lographic analysis.
the material presented near the crack tip. It is caused by stress concen Tensile tests were conducted to obtain the mechanical properties of
trations that may be due to cracks, defects, etc. It has been reported that the material for fracture toughness test and FE analysis. According to
the CTPZ has an important influence on the fracture behavior of the GB/T 228–2010, standard dog bone specimens with a gauge length of
material [29]. Finally, the correlation of CTPZ volume with load at 27 mm and diameter of 5 mm were designed, as shown in Fig. 2(a). The
different temperatures was obtained. loading axis was parallel to the BD at a tensile rate of 1.5 mm/min at four
temperatures (293, 223, 173, 123 K). Tensile tests were performed using
2. Specimen preparation and test procedure a microcomputer-controlled LD26.105 electronic universal testing ma
chine, with an electronic extensometer attached to the specimen to
2.1. Material and LPBF process procedure monitor the deformation of the gauge section. The axial strain was
automatically calculated by the computer based on the dimensions of
In order to perform the tensile and fracture toughness tests, the the tensile specimen. Due to the limitations of the test equipment, the
lowest test temperature was set at 123 K.
2
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Table 1
Nominal chemical compositions of the Ti6Al4V alloy powders (wt.%).
Element Ti Al V Fe O C N H
Table 2
LPBF processing parameters of the tested specimens.
Powder size Laser power Scanning speed Scanning space Layer thickness Preheating temperature Relative density Scanning strategy
[mm] [W] [mm/s] [mm] [mm] [◦ C] [%]
Fig. 1. Schematic representation of (a) the "cross-hatching" laser scanning strategy, (b) the orientation of the tensile and fracture specimens relative to the
build direction.
Fig. 2. Geometry and schematic description of (a) the tensile specimen, (b) the standard 1/2 T specimen (dimension in mm), (c) fracture toughness test setup.
Fig. 2(b) shows the 1/2 compact tension (CT) specimen (1/2 T) tensile testing machine automatically recorded the applied load using a
designed according to ASTM E399-20 for the determination of KIC [31]. force sensor. Liquid nitrogen spray was used for cooling, and the test was
Fatigue pre-cracking was performed on an electrohydraulic servo fa conducted in the environmental chamber device (WANCE, GD200A), as
tigue testing machine (EHF-EV01KZ-040-1A), and the crack length ratio shown in Fig. 2(c). Prior to loading, the specimen was kept at the cor
(a/W) of all specimens was 0.55 (a - crack length, W - width of the responding test temperature for at least 30 min. Then, the specimens
√̅̅̅̅
specimen). The load stress intensity range ΔK was 10 ~ 15 MPa m in a were fractured at a loading rate of 0.2 mm/min. Finally, the fracture
sinusoidal waveform at a frequency of 10 Hz, and the load ratio R was surfaces of the tensile and fracture specimens were observed by tungsten
constant at 0.1. A clip gauge (Epsilon, 3541C-006 M-040 M-LT) attached filament scanning electron microscopy (SEM, JSM-IT300). The surface
to the specimen was used to obtain the crack mouth opening displace roughness of all specimens was characterized by Ra according to GB/
ment (CMOD). Tensile load was applied to the 1/2 T specimen via two T131-2006. Ra is the arithmetic mean height of the profile lines. The
load pins inserted into the loading holes, and the WANCE TSE-104C-TS surface of the tensile specimens was polished to Ra 0.8. The main surface
3
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
of the 1/2 T specimens was polished to Ra 0.4, and the surface of the two the small-scale yielding (SSY) theory considering the plastic zone is
loading holes was polished to Ra 0.2. The experiment and specimen discussed in detail for corrections based on the linear elastic fracture
matrix are listed in Table 3. mechanics [37]. By considering the stress relaxation caused by plastic
deformation, the plastic zone in the crack tip is approximated as
( )2
2.3. Crack tip plasticity and finite element modeling 1 KI
rp = for plane stress (6)
π σ YS
2.3.1. Crack tip plasticity
( )2
The Mode I crack-tip plastic zone shapes estimated from the Irwin’s (1 − 2ν)2 KI
rp = for plane strain (7)
approach and the von Mises yield criterion gives the following formulas π σYS
[32]
In the case of SSY conditions (CTPZ is very small relative to the
( )2 [ ]
1 KI 3 characteristic length L, e.g., crack size, a), both KI and J-integral char
ry (θ) = 1 + cosθ + sin2 θ for plane stress (2)
4π σYS 2 acterize the crack tip conditions [32]. At a short distance from the crack
√̅̅
tip, defined as rk (rk ≈ 0.3a ∼ 0.5a), the stress is proportional to 1/ r.
( )2 [ ]
1 KI 3 By assuming monotonic, quasi-static loading, a J-dominated region oc
ry (θ) = (1 − 2ν)2 (1 + cosθ) + sin2 θ for plane strain (3)
4π σYS 2 curs in the plastic zone rp (rp ≪rk , at least one order of magnitude), where
the elastic singularity no longer applies. Well within the plastic zone, the
where KI is the stress intensity factor which is used to characterize the HRR solution is approximately valid and stresses vary as r− 1/(n+1) . The
linear elastic stress field at the crack tip [33], θ is the angle with the finite strain region occurs within the plastic zone rp about 2δ (δ is the
crack plane in polar coordinates, ν is Poisson’s ratio. However, Eqs. (2) crack tip opening displacement proposed by Wells [38]) of the crack tip,
and (3) are not strictly correct in this case because they are based on a where large deformations invalidate the HRR theory. In summary, for
purely linear elastic analysis. SSY theory, KI uniquely characterizes the crack tip conditions, although
Rice [29] proposed a fracture parameter of J-integral and showed √̅̅
the 1/ r singularity does not exist all the way to the crack tip, and the
that for deformation plasticity (nonlinear elasticity), the value of J-in typical KI-dominated region is rp < r≪rk ; Similarly, the J-integral
tegral is independent of the integration path around the crack. Using the uniquely characterizes the crack-tip conditions, although the deforma
concept of J-integral, Hutchinson [34] and Rice and Rosengren [35] tion plasticity and small strain assumptions are invalid in the finite
independently evaluated the character of crack tip fields in a nonlinear strain regime. Under SSY conditions, the J-integral is related to the KI as
elastic power-law hardening material. When elastic strains are included,
the relationship between plastic strain and stress for uniaxial deforma J=
(1 − ν2 )KI2
(8)
tion is given by E
σ
( )
σ0 σ n Therefore, we used the elastic–plastic J-integral to characterize the
ε = + α (4) crack tip filed and to obtain the CTPZ in the SSY condition instead of
E E σ0
correcting the linear elastic fracture mechanics KI in this study.
where σ 0 is the reference stress value, usually equal to the yield strength
of the material, E is the Young’s modulus, α is a dimensionless constant, 2.3.2. Finite element modeling
and n is the strain-hardening exponent. Eq. (4) is known as the Ramberg- FE analysis was used to evaluate the effects of CTPZ on the fracture
Osgood equation, and is widely used for curve-fitting of true stress– toughness of 1/2 T specimens tested at different temperatures based on
strain data. Hutchinson obtained the following asymptotic variation of the contour integral using the commercial FE analysis software ABA
stress and strain ahead of the crack tip QUS. The complex microstructure of the LPBF Ti6Al4V alloy was
( ) ignored, and the anisotropy was not considered. Since the specimens
σ ij = σ0
J 1 ∼
n+1 σ (n, θ) (a) were heat treated, the material property of LPBF Ti6Al4V was assumed
ij
ασ0 ε0 In r as isotropic and elastoplastic, and the Poisson’s ratio v = 0.34 and
( ) (5)
J n ∼ Young’s modulus were given in Table 4. The true stress–strain curves
εij = αε0 r n+1 εij (n, θ) (b) were used to consider the plastic property of the material in this study, as
ασ0 ε0 In r
shown in Fig. 6. Taking advantage of geometric symmetry, one quarter
where ε0 = ε0 /E, In is an integration constant that depends on n, σ ij and
∼ of the 1/2 T specimen was modeled. On the fracture surface (YOZ (x = 0)
plane), the X-axis symmetry constraint (Ux = URy = URz = 0) was set;
ε ij are the dimensionless functions of n and θ. Rice and Rosengren ob
∼
on the mid-symmetry surface (XOY (z = 0) plane) of the specimen, the Z-
tained essentially identical results to Hutchinson’s solutions for plane
axis symmetry constraint (Uz = URy = URx = 0) was used as shown in
strain in a different format. Eq. (5) is therefore referred to as the HRR
Fig. 3(a).
field, and the J-integral is used to describe the singularity intensity of the
The boundary conditions used in the finite element calculation are
crack-tip stress field for elastic–plastic hardening materials [36].
important to ensure that the deformation response is consistent with the
In fact, most metallic materials have a plastic zone at the crack tip
macroscopic mechanical behavior of the material. In a three-
before and during fracture, and the crack is also blunted by plastic
dimensional fatigue crack propagation simulation, it is shown that the
deformation. Therefore, within a small local area, plastic yielding will
horizontal displacement constraints in all boundary conditions are in
make the theoretical results based on the linear elastic response unable
better agreement with the analytical results than applying the load to
to describe the stress and strain behavior in this local area. The effect of
the reference point coupled to the loading region [39]. For convergence
and accuracy considerations, the horizontal displacement constraint of
Table 3 0.3 mm was applied to a rigid pin in contact (frictionless) with the
Summary of different experiment types and number of specimens with respect to
specimen instead of applying loads in this study. The element type used
the test temperatures in each case.
was a 20-node quadratic hexahedral reduced integration element
Test temperatures [K] 293 223 173 123 (C3D20R brick element). Considering the influence of stress singularity,
Experiment type Specimen type Test number the brick element was converted to a 15-node quadratic triangular prism
Monotonic tensile Tensile specimen 3 2 2 2 element (C3D15 wedge element) at the crack tip of the specimen, thus
Fracture test 1/2 T specimen 4 3 3 3 avoiding the use of additional singularity elements in the finite element
4
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Table 4
Tensile property of LPBF Ti6Al4V alloy at different test temperatures. Properties of wrought Ti6Al4V alloys at different temperatures are taken from [28] and [44].
Manufacturing Technology Temp. σYS [MPa] σUTS [MPa] E εf [%] n
[K] [GPa] [/]
Fig. 3. (a) Boundary conditions and (b) crack tip mesh details of the 3D 1/2 T specimen.
simulation. Due to the severe plastic deformation in the crack tip region,
a refined mesh was performed to improve the calculation accuracy as
shown in Fig. 3 (b). The number of elements in the model was 23080 and
the J-integral was calculated using the ABAQUS procedure.
Fig. 4. Initial microstructure of the LPBF Ti6Al4V alloy: EBSD inverse pole figure map in the TD-RD plane (a) and the TD-AD plane (b), the {0001} pole figure of the
TD-RD plane (c) and the TD-AD plane (d). The white dashed lines in (a) and (b) represent prior β grain boundaries.
5
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
grains remain unchanged [9]. Additionally, the relatively rapid cooling wrought Ti6Al4V alloys, the εf at 20 K is higher than that at 77 K. This is
resulting from the single-step cooling process leads to the formation of because that the plastic deformation of the Ti6Al4V alloy changes from
the basket weave structure frequently observed in LPBF Ti6Al4V alloy slip to twinning-dominated flow when the temperature is lowered from
[17]. It is apparent that the material predominantly consists of a room temperature [37]. Overall, the effect of temperature on the tensile
lamellar α structure, as evidenced by needle-shaped α laths within the β properties of the LPBF Ti6Al4V alloy is similar to that of the conven
grains. tional wrought Ti6Al4V alloys at 77–293 K. That is, the tensile strength
The maximum peak intensity, which is also known as texture increases and the elongation decreases with decreasing temperature
strength of the {0001} pole figure (PF) of the TD-RD plane and the TD- [28,44], as shown in Table 4. The σ YS at 293 K is 341 MPa higher than
AD plane are 7.36 and 15.69 respectively, as shown in Fig. 4(c-d). With that of the specimen at 123 K. The Young’s modulus increases gradually
almost the same number of variants, the texture strength differs by with the decreasing of temperature, but the Young’s modulus of the
almost 50%, indicating that there are more laths with similar orienta specimen at 123 K is lower than that at 173 K. Considering the number of
tions in the TD-RD plane. It is unlikely that the anisotropy can be specimens and the accuracy of test, more experiments are still needed to
completely eliminated by stress relieving annealing. It is expected that investigate the relationship between temperature and Young’s modulus.
microstructure anisotropy has significant effect on mechanical proper Overall, σ UTS is not much higher than σYS at the same temperature. This
ties such as the tensile strength, high-cycle and very-high-cycle fatigue may be related to the low work- hardening capability of the LPBF
properties, and crack growth rate [40,41]. The columnar PBGs structure
may be responsible for the anisotropy observed in the LPBF Ti6Al4V
alloy [42]. The heat treatment process significantly affects the texture of
LPBF Ti6Al4V, which in turn affects the mechanical properties. Overall,
the proper heat treatment not only mitigates the anisotropy in proper
ties, but also improves the ductility, fracture toughness, and fatigue
crack growth resistance [24].
Fig. 5 shows the optical microscopy images of the LPBF Ti6Al4V
alloy. The prior β grains in the TD-RD plane show a checkerboard
pattern and the average grain length is about 90 μm, which is the same
as the laser scanning space, as shown in Fig. 5(a). The microstructure of
the TD-AD plane in Fig. 5(b) shows the vertical columnar prior β grains
which are aligned along the BD. These grains are discontinuous due to
the mismatch between the successive laser tracks. Kumar et al. [16]
argued that due to the existence of the melt pool boundaries (MPBs),
columnar grains in the TD-AD plane become shorter and grain size be
comes smaller. In addition, only a small number of defects were
observed in the LPBF Ti6Al4V alloy compared to the as-built Ti6Al4V
alloy [43], indicating that heat treatment can effectively reduce the
porosity and improve the relative density.
Fig. 6. True stress–strain curves of LPBF Ti6Al4V alloy.
Fig. 5. Optical microscopy images of the LPBF Ti6Al4V alloy: (a) TD-RD plane, (b) TD-AD plane.
6
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
7
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Fig. 8. Representative fracture surfaces of the tensile specimens: (a-b) 293 K, (c-d) 223 K, (e-f) 173 K, (g-h) 123 K. The red and black arrows in (b-d) indicate the
defect responsible for the onset of fracture and the location of plateaus or basins, respectively.
8
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Fig. 9. The force-CMOD curves (a) and KIC values (b) of 1/2 T specimens at different temperatures.
Table 5
Fracture toughness of the tested specimens at different temperatures (KIC are the average values of the valid KQ ).
Temperature [K] Specimen a/W KQ Criterion 1 Criterion 2 KIC rp [mm]
√̅̅̅̅̅ √̅̅̅̅̅
number [/] [MPa m] [MPa m]
perpendicular to the build direction are clearly shown at the junction of colonies, the crack will appear deflection. According to [52], the
the two regions are clearly shown. As the temperature decreases, the translamellar crack is strongly dependent on the active slip planes, while
ravines become smaller and the fracture surface of the unstable crack the interlamellar crack is closely related to the misorientation between
propagation region becomes flatter. Usually, larger ravines and cone- the adjacent α-lamellas. In addition, the large orientation deviation
shaped hills are caused by crack deflection during fracture and more between α lamellar will promote the deflection of cracks or lead to the
energy is consumed [48], which may be a reason for the increased alternating propagation of interlamellar cracks and translamelar cracks.
fracture toughness at higher temperature. Eventually, a serrated crack propagation path is formed. In general, the
The main crack initiations observed in the LPBF Ti6Al4V alloy are: crack propagation path of the specimen is relatively flat, indicating that
smooth facets with small cracks, LOF regions with un-melted powder there is a small plastic deformation, which is consistent with the duc
and gas pores, as shown in Fig. 11. According to [49], defects are usually tile–brittle mixed fracture mode of the low temperature tensile specimen
caused by insufficient energy (LOF defects) or excessive energy (gas shown in Fig. 8(c).
pores) in the metallic LPBF process. In this study, both excessive energy
and insufficient energy defects appeared in the fracture surfaces. 3.3.2. Temperature effect on crack tip plastic zone
α-phase facet defects and microscopic small cracks are reported to be the A significant difference can be observed in the plastic deformation
source of crack initiation in traditionally manufactured Ti6Al4V alloy region of 1/2 T specimens at different temperatures, as shown in Fig. 10.
[50]. However, cracks in LPBF Ti6Al4V alloy can also originate from To clearly get the size and shape of the CYPZ at different temperatures,
pores and LOF defects. That is, there are several competing fracture finite element analysis (FEA) based on the elastic–plastic J-integral was
mechanisms in additively manufactured titanium alloys. The effect of used to characterize the crack tip filed. Von Mises stress distributions of
defects on the fracture mechanism is significant, but defects in LPBF the specimens at different temperatures are shown in Fig. 13. The gray
parts can be minimized by the use of high-quality powders, substrate area near the crack tip in the stress contour represents the plastic zone,
preheating, optimization of process parameters and post heat treatment where the von Mises stress exceeds the yield stress of the material. The
[51], resulting in improved fracture toughness. The final fracture size of the plastic zone is relatively small compared to the crack length
toughness of additively manufactured Ti6Al4V alloy is ultimately and other characteristic dimensions. It can also be seen that the plastic
determined by the number and location of defects within the specimen. yield occurs only near the crack tip position, and the plastic zone has a
Fig. 12 illustrates the interaction between the crack path and the α classic butterfly shape in the XOY (z = 0) plane. As the temperature
colonies (prior β grains) characterized by EBSD of the sample at 173 K. decreases, the maximum of the von Mises stress increases and the plastic
The result shows that the primary crack propagates through the α col zone of the crack tip gradually decreases. This may be the reason for the
onies, namely translamellar propagation, and at the junction of the reduced fracture toughness of the material at low temperature, since the
9
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Fig. 10. Representative fracture surfaces of 1/2 T specimens at different temperatures: (a) 293 K, (b) 223 K, (c) 173 K, (d) 123 K.
Fig. 11. Typical defects in the LPBF Ti6Al4V alloy: (a) smooth facets with small cracks, (b) LOF defects with un-melted powder particles, (c) gas pores.
plastic zone is the result of plastic deformation, which is accompanied by temperature. The plastic zone calculated by FEA is closer to the plane
drastic energy consumption. strain plastic zone. It indicates that our finite element simulation sat
Fig. 14 shows the comparisons for the CTPZ contour lines in the mid- isfies the plane strain condition, which is consistent with the results of
symmetry plane between the results obtained by FEA and those pre most specimens meeting the KIC validity criteria, and also proves the
dicted by the theoretical solutions Eqs. (2) and (3) when the horizontal reliability of the FEA results in this study.
displacement constraint reaches to the limit of 0.3 mm at different Fig. 15(a) shows the distributions of the J-integral in the mid-
temperatures. In the mid-symmetry plane, the plane strain plastic zone is symmetry plane. The J-integral increases monotonically with
much smaller than the plane stress plastic zone, the former condition increasing loading displacement. It is interesting to notice that the J-
suppresses yielding, resulting in a smaller plastic zone for a given load integral of the specimen at 173 K is larger than that of 123 K for the same
value. The area of the plastic zone decreases with the decreasing of load. It can be found that the variation of J-integral with temperature
temperature. The FEA plastic zone area of the specimen at 123 K is only follows the same trend as the variation of Young’s modulus with tem
about one-third of that at 293 K. The smaller the plastic zone is, the perature. A subroutine was developed to accurately calculate the vol
specimen at low temperature is less likely to yield and more prone to ume of the plastic zone (Vp) near the crack tip, as shown in Fig. 15(b).
brittle fracture than the specimen at room temperature. This explains When the loading displacement is less than 0.1 mm, the specimen does
the increase in tensile yield strength at low temperature, and is consis not undergo plastic deformation, and the Vp is 0. For the same load
tent with the mixed ductile–brittle fracture mode exhibited at low displacement, the Vp decreases with the decreasing of temperature.
10
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Fig. 12. EBSD IPF map of the crack path for a 173 K sample. The white dotted lines indicate α colonies and the yellow dotted lines mark the crack path profile.
Fig. 13. Von Mises stress distributions of 1/2 T specimens at different temperatures: (a) 293 K, (b) 223 K, (c) 173 K, (d) 123 K.
Researchers have found that both high-purity Ti and Ti-5at%Al have [53]. For LPBF Ti6Al4V alloy, the dislocation motion can also be
significantly harder dislocation motion at liquid nitrogen temperatures retarded at low temperature, which resulting in the reduction of plastic
(77 K) than at room temperature. At low temperatures, the dislocation energy dissipation. The macroscopic manifestation is that the fracture
motion of high-purity Ti is even more difficult than that of Ti-5at%Al toughness decreases with the decreasing of temperature. In some
11
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Fig. 14. Results predicted by theoretical solutions Eqs. (2) and (3) compared with FEA for the CTPZ contour lines in the mid-symmetry plane (z = 0) of 1/2 T
specimens at various temperatures.
Fig. 15. The distribution of the J-integral in the mid-symmetry plane (a) and the volume of the plastic zone near the crack tip (b) at different temperatures.
12
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
Correia: Writing – review & editing. Youshi Hong: Writing – review &
editing.
Data availability
Acknowledgements
References
temperature. The J-integral is affected by both loading and material
properties. In the future, it may be possible to find the unified rela [1] J.H. Martin, B.D. Yahata, J.M. Hundley, J.A. Mayer, T.A. Schaedler, T.M. Pollock,
tionship between Vp and J-integral at different temperatures, so that the 3D printing of high-strength aluminium alloys, Nature 549 (2017) 365–369.
fracture toughness of materials at different temperatures can be pre [2] Z. Snow, R. Martukanitz, S. Joshi, On the development of powder spreadability
metrics and feedstock requirements for powder bed fusion additive manufacturing,
dicted by FEA. However, such an attempt would require a large number Addit. Manuf. 28 (2019) 78–86.
of experiments with FEA results to ensure reliability. [3] I. Gibson, D. Rosen, B. Stucker, M. Khorasani, Additive Manufacturing Technology,
in, Springer Nature Switzerland AG, Switzerland, 2021.
[4] W.E. Frazier, Metal Additive Manufacturing: A Review, J. Mater. Eng. Perform. 23
4. Conclusions (2014) 1917–1928.
[5] J.J. Lewandowski, M. Seifi, Metal Additive Manufacturing: A Review of Mechanical
In this paper, effects of temperature on the tensile and fracture Properties, Annu. Rev. Mat. Res. 46 (2016) 151–186.
[6] L. Du, G. Qian, L. Zheng, Y. Hong, Influence of processing parameters of selective
performance of the LPBF Ti6Al4V alloy were investigated. The micro laser melting on high-cycle and very-high-cycle fatigue behaviour of Ti-6Al-4V,
structure anisotropy and fracture surfaces of specimens at different Fatigue Fract. Eng. Mater. Struct. 44 (2020) 240–256.
temperatures were analyzed. The following conclusions are drawn: [7] N.T. Aboulkhair, N.M. Everitt, I. Ashcroft, C. Tuck, Reducing porosity in AlSi10Mg
parts processed by selective laser melting, Addit. Manuf. 1–4 (2014) 77–86.
[8] G. Kasperovich, J. Hausmann, Improvement of fatigue resistance and ductility of
(1) The tensile strength obviously increases with decreasing tem TiAl6V4 processed by selective laser melting, J. Mater. Process. Technol. 220
perature. The yield strength of the specimen at 123 K is 35% (2015) 202–214.
higher than that at 293 K, and the corresponding fracture elon [9] G.M. Ter Haar, T.H. Becker, Selective Laser Melting Produced Ti-6Al-4V: Post-
Process Heat Treatments to Achieve Superior Tensile Properties. Materials (Basel)
gation decreases by only 2.9%. The fracture mode changes from 2018; 11.
ductile fracture at room temperature to a mixed mode at low [10] U. Tradowsky, J. White, R.M. Ward, N. Read, W. Reimers, M.M. Attallah, Selective
temperature. laser melting of AlSi10Mg: Influence of post-processing on the microstructural and
tensile properties development, Mater. Des. 105 (2016) 212–222.
(2) A correlation of fracture toughness and temperature between 123 [11] S. Leuders, M. Thöne, A. Riemer, T. Niendorf, T. Tröster, H.A. Richard, H.J. Maier,
and 293 K was obtained. The fracture toughness of the specimen On the mechanical behaviour of titanium alloy TiAl6V4 manufactured by selective
at 123 K is 16.2% lower than that at 293 K. Smooth facets with laser melting: Fatigue resistance and crack growth performance, Int. J. Fatigue 48
(2013) 300–307.
small cracks, LOF regions and gas pores can form the crack [12] T.H. Becker, P. Kumar, U. Ramamurty, Fracture and fatigue in additively
initiation in LPBF Ti6Al4V alloy. manufactured metals, Acta Mater. 219 (2021).
(3) The crack tip plastic zone size decreases with the decreasing of [13] L. Zhu, N. Li, P.R.N. Childs, Light-weighting in aerospace component and system
design, Propul. Power Res. 7 (2018) 103–119.
temperature. This may be because the dislocation motion is [14] D.D. Gu, W. Meiners, K. Wissenbach, R. Poprawe, Laser additive manufacturing of
retarded at low temperature, resulting in the reduction of plastic metallic components: materials, processes and mechanisms, Int. Mater. Rev. 57
energy dissipation. The variation of plastic zone is consistent with (2013) 133–164.
[15] L. Thijs, F. Verhaeghe, T. Craeghs, J.V. Humbeeck, J.-P. Kruth, A study of the
that of fracture toughness with temperature.
microstructural evolution during selective laser melting of Ti–6Al–4V, Acta Mater.
(4) The LPBF Ti6Al4V alloy exhibits strong microstructural anisot 58 (2010) 3303–3312.
ropy. The texture strength of the samples differs by nearly 50% in [16] P. Kumar, O. Prakash, U. Ramamurty, Micro-and meso-structures and their
the direction parallel to and perpendicular to the build direction. influence on mechanical properties of selectively laser melted Ti-6Al-4V, Acta
Mater. 154 (2018) 246–260.
This is determined by the micro- and meso-structures of the LPBF [17] J. Yang, H. Yu, J. Yin, M. Gao, Z. Wang, X. Zeng, Formation and control of
Ti6Al4V alloy, and stress relieving annealing cannot completely martensite in Ti-6Al-4V alloy produced by selective laser melting, Mater. Des. 108
eliminate the anisotropy. (2016) 308–318.
[18] L. Thijs, M.L. Montero Sistiaga, R. Wauthle, Q. Xie, J.-P. Kruth, J., Van Humbeeck,
Strong morphological and crystallographic texture and resulting yield strength
CRediT authorship contribution statement anisotropy in selective laser melted tantalum, Acta Mater. 61 (2013) 4657–4668.
[19] S.L. Lu, Z.J. Zhang, R. Liu, X.H. Zhou, X.G. Wang, B.N. Zhang, X.M. Zhao, J. Eckert,
Z.F. Zhang, Optimal tensile properties of laser powder bed fusion hereditary
Yingmeng Xiao: Investigation, Methodology, Software, Formal basket-weave microstructure in additive manufactured Ti6Al4V, Addit. Manuf. 59
analysis, Data curation, Writing – original draft, Writing – review & (2022).
editing. Guian Qian: Conceptualization, Methodology, Supervision, [20] P. Andresen, B. Antolovich, ASM Handbook, in: Fatigue and Fracture, ASM
International, vol. 19, 1996, p. 1057.
Writing – review & editing. Jingyu Sun: Conceptualization, Writing – [21] F. Bartolomeu, M. Gasik, F.S. Silva, G. Miranda, Mechanical Properties of Ti6Al4V
review & editing. Filippo Berto: Writing – review & editing. Jose A.F. Fabricated by Laser Powder Bed Fusion: A Review Focused on the Processing and
Microstructural Parameters Influence on the Final Properties. Metals 2022;12.
13
Y. Xiao et al. Theoretical and Applied Fracture Mechanics 125 (2023) 103931
[22] V. Cain, L. Thijs, J. Van Humbeeck, B. Van Hooreweder, R. Knutsen, Crack [40] W. Sun, Y.e. Ma, W. Huang, W. Zhang, X. Qian, Effects of build direction on tensile
propagation and fracture toughness of Ti6Al4V alloy produced by selective laser and fatigue performance of selective laser melting Ti6Al4V titanium alloy, Int. J.
melting, Addit. Manuf. 5 (2015) 68–76. Fatigue 2020; 130.
[23] S. Cao, Y. Zou, C.V.S. Lim, X. Wu, Review of laser powder bed fusion (LPBF) [41] G. Qian, Y. Li, D.S. Paolino, A. Tridello, F. Berto, Y. Hong, Very-high-cycle fatigue
fabricated Ti-6Al-4V: process, post-process treatment, microstructure, and behavior of Ti-6Al-4V manufactured by selective laser melting: Effect of build
property. Light: Advanced Manufacturing 2021;2. orientation, Int. J. Fatigue 136 (2020).
[24] P. Kumar, U. Ramamurty, Microstructural optimization through heat treatment for [42] Y. Kok, X.P. Tan, P. Wang, M.L.S. Nai, N.H. Loh, E. Liu, S.B. Tor, Anisotropy and
enhancing the fracture toughness and fatigue crack growth resistance of selective heterogeneity of microstructure and mechanical properties in metal additive
laser melted Ti-6Al-4V alloy, Acta Mater. 169 (2019) 45–59. manufacturing: A critical review, Mater. Des. 139 (2018) 565–586.
[25] M.R. Khosravani, F. Berto, M.R. Ayatollahi, T. Reinicke, Fracture behavior of [43] N. Sanaei, A. Fatemi, Defects in additive manufactured metals and their effect on
additively manufactured components: A review, Theor. Appl. Fract. Mech. 109 fatigue performance: A state-of-the-art review, Prog. Mater Sci. 117 (2021).
(2020). [44] R. Sołtysiak, J. Małecka, The analysis of cyclic properties of Ti-6Al-4V titanium
[26] G. Singh, U. Ramamurty, Reprint: Boron modified titanium alloys, Prog. Mater Sci. alloy at room and liquid nitrogen temperature, Mater. Sci. Eng. A 734 (2018)
120 (2021). 476–484.
[27] T. Yuri, Y. Ono, T. Ogata, Notch effects on high-cycle fatigue properties of [45] A. Soltani-Tehrani, M. Habibnejad-Korayem, S. Shao, M. Haghshenas, N. Shamsaei,
Ti–6Al–4V ELI alloy at cryogenic temperatures, Cryogenics 46 (2006) 30–36. Ti-6Al-4V powder characteristics in laser powder bed fusion: The effect on tensile
[28] G. Singh, G. Bajargan, R. Datta, U. Ramamurty, Deformation and strength of and fatigue behavior, Addit. Manuf. (2022) 51.
Ti–6Al–4V alloyed with B at cryogenic temperatures, Mater. Sci. Eng. A 611 (2014) [46] M. Simonelli, Y.Y. Tse, C. Tuck, Effect of the build orientation on the mechanical
45–57. properties and fracture modes of SLM Ti–6Al–4V, Mater. Sci. Eng. A 616 (2014)
[29] J.R. Rice, A Path Independent Integral and the Approximate Analysis of Strain 1–11.
Concentration by Notches and Cracks, J. Appl. Mech. 35 (1968) 379–386. [47] Y. Lu, Y.H. Zhang, E. Ma, W.Z. Han, Relative mobility of screw versus edge
[30] B. Zhang, Y. Li, Q. Bai, Defect Formation Mechanisms in Selective Laser Melting: A dislocations controls the ductile-to-brittle transition in metals, Proc. Natl. Acad.
Review, Chin. J. Mech. Eng. 30 (2017) 515–527. Sci. U S A 118 (2021).
[31] W. Conshohocken, Standard Test Method for Linear-Elastic Plane-Strain Fracture [48] H. Wang, Q. Zhao, S. Xin, Y. Zhao, W. Zhou, W. Zeng, Microstructural morphology
Toughness of Metallic Materials, in, PA, 2020. effects on fracture toughness and crack growth behaviors in a high strength
[32] T.L. Anderson, Fracture Mechanics Fundamentals And Applications 3rd Ed, in, CRC titanium alloy, Mater. Sci. Eng. A 821 (2021).
Press, New York, 1995. [49] G. Kasperovich, J. Haubrich, J. Gussone, G. Requena, Correlation between porosity
[33] G.R. Irwin, Analysis of Stresses and Strains Near the End of a Crack Traversing a and processing parameters in TiAl6V4 produced by selective laser melting, Mater.
Plate, J. Mech. Phys. Solids 24 (1957) 361–364. Des. 105 (2016) 160–170.
[34] J.W. Hutchinson, Singular behaviour at the end of a tensile crack in a hardening [50] S. Tammas-Williams, P.J. Withers, I. Todd, P.B. Prangnell, The Influence of
material, J. Mech. Phys. Solids 16 (1968) 13–31. Porosity on Fatigue Crack Initiation in Additively Manufactured Titanium
[35] J.R. Rice, G.F. Rosengren, Plane Strain Deformation near a Crack Tip in a Power- Components, Sci. Rep. 7 (2017) 7308.
Law Hardening Material, J. Mech. Phys. Solids 16 (1968) 1 -+. [51] W.J. Sames, F.A. List, S. Pannala, R.R. Dehoff, S.S. Babu, The metallurgy and
[36] X.-K. Zhu, J.A. Joyce, Review of fracture toughness (G, K, J, CTOD, CTOA) testing processing science of metal additive manufacturing, Int. Mater. Rev. 61 (2016)
and standardization, Eng. Fract. Mech. 85 (2012) 1–46. 315–360.
[37] S. Diiorio, L. Briottet, E. Rauch, D. Guichard, Plastic deformation, damage and [52] S. Birosca, J.Y. Buffiere, F.A. Garcia-Pastor, M. Karadge, L. Babout, M. Preuss,
rupture of PM Ti–6Al–4V at 20K under monotonic loading, Acta Mater. 55 (2007) Three-dimensional characterization of fatigue cracks in Ti-6246 using X-ray
105–118. tomography and electron backscatter diffraction, Acta Mater. 57 (2009)
[38] A.A. Wells, Application of fracture mechanics at and beyond general yielding. 5834–5847.
British Welding Research Assoc Rep M13/63 1963;10:563-570. [53] B. Zhao, P. Huang, L. Zhang, S. Li, Z. Zhang, Q. Yu, Temperature Effect on Stacking
[39] H. Xin, J.A.F.O. Correia, M. Veljkovic, Three-dimensional fatigue crack Fault Energy and Deformation Mechanisms in Titanium and Titanium-aluminium
propagation simulation using extended finite element methods for steel grades Alloy, Sci. Rep. 10 (2020) 3086.
S355 and S690 considering mean stress effects, Eng. Struct. 227 (2021). [54] D.-K. Chang, Q.-Q. Wang, H. Chen, X.-C. Zhang, J.-F. Wen, S.-T. Tu, Effects of
defect size and location on high cycle fatigue life of a maraging stainless steel at
ambient and cryogenic temperatures, Int. J. Fatigue 161 (2022).
14